Journal of Human Hypertension (1998) 12, 807–810  1998 Stockton Press. All rights reserved 0950-9240/98 $12.00 http://www.stockton-press.co.uk/jhh FOR DEBATE Beta-blockers for hypertension: time to call a halt

DG Beevers University Department of Medicine, City Hospital, Birmingham B18 7QH, UK

Beta-blockers are not very effective at lowering blood the endorsement of beta-blockers by the British Hyper- pressure in elderly hypertensive patients or in Afro- tension Society and other guidelines committees, Caribbeans and these two groups represent a large pro- except possibly for severe resistant hypertension, high portion of people with raised blood pressure. Further- risk post-infarct patients and those with angina pectoris. more they do not prevent more heart attacks than the The time has come to move across to newer, safer, more thiazide diuretics. Beta-blockers can also be dangerous tolerable and more effective antihypertensive agents in many hypertensive patients and even when these whilst continuing to use thiazide diuretics in low doses drugs are not contraindicated, they cause subtle and in the elderly as first choice, providing there are no depressing side effects which should preclude their contraindications. usefulness. The time has come therefore to reconsider

Keywords: beta-blockers; hypertension

Introduction Safety and tolerability

Beta-adrenergic blockers were first introduced in the There is little doubt that the beta-blockers are the early 1960s for the treatment of angina pectoris. most unsafe of all antihypertensive drugs. They can Their antihypertensive properties were not fully precipitate or worsen heart failure in patients with recognised until the celebrated paper by Pritchard myocardial damage and they are contraindicated in and Gillam in 1964.1 They rapidly became popular patients with asthma. It is not uncommon for because they had fewer side effects than the older patients to be admitted to hospital with either of medications like methyldopa, guanethidine, detriso- these conditions having being rendered acutely quine and bethanidine which were available at the unwell by the introduction of a beta-blocker for the time. Beta-blockers were used in several of the long- term morbidity and mortality trials in the treatment treatment of their hypertension. These drugs also of hypertension, either alone or in comparison with worsen Raynaud’s phenomenon and intermittent claudication and even peripheral gangrene has the thiazide diuretics. Despite their poor showing in 4 these trials, the guidelines committee of the British been reported. Hypertension Society and the sixth American Joint The other problem with the beta-blockers have National Committee have taken the view that beta- been their more subtle side effects on exercise toler- blockers, along with thiazides, are the preferred ance, sleep patterns and the capacity to concen- 5 option in the first line treatment of hypertension trate. Tiredness and lethargy are unacceptable side unless they are specifically contraindicated.2,3 effects in previously symptomless mild hyperten- Over the last 15 years several new classes of sives who need to take antihypertensive drugs for hypertensive drugs have become available which are the rest of their lives. Furthermore the beta-blockers also effective at lowering blood pressure and more have been shown to cause impotence.5 Perhaps the recently the angiotensin receptor antagonists have acid test of whether these side effects are acceptable been found to be almost devoid of side effects. is to ask a group of clinicians whether they them- As the various national and international guide- selves would choose to take a beta-blocker if they lines committees deliberate on their next set of re- had uncomplicated mild hypertension. Very few commendations there is now a good case for with- would say ‘yes’. Why then do they prescribe them drawing any endorsement of the beta-blockers as for their patients? either first or even second line antihypertensive It is probably true that the hydrophilic beta-block- drugs. ers like atenolol and bisoprolol have fewer side effects and it is also true that many beta-blockers are often being used at unnecessary high doses. How- ever, even when atenolol is prescribed at the dose of 50 mg daily, side effects still occur.6

Correspondence: Professor DG Beevers Beta-blockers for hypertension DG Beevers 808 Antihypertensive efficacy response of renin to sodium depletion. Elderly patients have low plasma renin levels, possibly due to In young hypertensive patients the beta-blockers are early nephrosclerosis and black patients have low not more effective than the other antihypertensive renin levels possibly related to reduced beta-adre- drugs and therefore have no particular advantages. nergic responsiveness, exaggerated nephrosclerosis or In older patients however, there is good evidence differences in sodium handling.15,16 Whatever the that beta-blockers are less effective than other drugs. mechanisms, these age and ethnic differences in anti- In the MRC trial of mild hypertension in the eld- hypertensive drug efficacy are well recognised and erly,7 the Coope and Warrender study8 and the not controversial. Swedish Trial of Old Persons (STOP),9 mildly hypertensive patients were given beta-blockers as their first-line antihypertensive agent. In the Beta-blockers as second line therapy majority it proved necessary to add in a thiazide Whilst the beta-blockers are logical add-in drugs in diuretic (Table 1). By contrast in the patients in the patients who are already receiving either a thiazide MRC trial who received a diuretic as first-line ther- diuretic or a calcium channel blocker, they provide apy and participants in the European Working Party little benefit when added to an angiotensin- on Hypertension in the Elderly (EWPHE) trial, converting enzyme (ACE) inhibitor or an angioten- where thiazide diuretics were also used as first-line sin receptor antagonist.17 As these drugs effectively therapy, add-in drugs were only necessary in a min- block the renin-angiotensin system there seems little ority of patients.10 point in further blocking it by giving beta-blockers In the treatment of isolated systolic hypertension, as there is no convincing evidence of genuine syn- Avanzini et al11 found that atenolol when used ergy. alone produced a good antihypertensive effect in less than 20% of patients whilst diuretics controlled blood pressure in almost half of the patients. This Long term outcome trials of cardiovascular latter figure is similar to that obtained with diuretics mortality and morbidity in the Systolic Hypertension in the Elderly Pro- The beta-blockers have been compared with the thi- gramme (SHEP).12 As hypertension, and particularly azide diuretics to investigate their capacity to pre- isolated systolic hypertension, are largely diseases vent heart attacks and strokes in four studies, two of the elderly and beta-blockers are barely effective of which were organised by the Medical Research in these patients, there is a very good reason for not Council.7,18–20 In the MRC studies the beta-blockers using them at all. More sensible drugs to use are the had no significant impact overall on the prevention thiazide diuretics or the calcium channel blockers. of coronary heart disease and had less impact on the A survey of general practitioner prescribing habits stroke than the thiazide diuretics (Table 2). It was in elderly patients has shown that the thiazides are suggested that a subgroup of patients in the MRC the most popular option.13 trial in younger patients, who were non-smokers, might have benefited from beta-blockade but this African origin patients subgroup analyses itself must be treated with con- siderable caution.20 A similar trend was found in the Several studies in the United States of America have IPPPSH study18 but no such trend was found in demonstrated that beta-blockers are not very effec- either the HAPPHY study19 or its subgroup, the tive in patients of African origin.14 As hypertension MAPHY study.21 Indeed in this latter study, beta- is about twice as common in black people than in blockers appeared to be more effective in preventing whites and again bearing in mind most hyperten- heart attacks in smokers. By sharp contrast in the sives are elderly there is a very good reason for never SHEP10 and the EWPHE12 studies where thiazide employing beta-blockers in black patients. Again diuretics were used as first-line therapy, coronary more sensible drugs are the thiazide diuretics or the prevention rates were impressive. In the Coope and calcium channel blockers. Warrender8 study where beta-blockers were used, This poor response to beta-blockers is related to the the reduction of coronary heart disease was smaller finding that these drugs tend to be less effective in than in all the other elderly hypertension trials. people with low plasma renin levels and an impaired Beta-blockers therefore do no prevent heart attacks and barely prevent strokes in hypertensive patients. Table 1 The proportion (%) of elderly patients whose blood pressures were controlled on their first-line drugs, or who did not require the addition of second line drugs in the five long term Post infarction studies outcome trials of the treatment of hypertension which employed either thiazide diuretics or beta-blockers. For abbreviations see Several studies have shown that beta-blockade is text effective at the secondary prevention of myocardial infarction in high risk patients.22 Despite this only Thiazide diuretic Beta-blocker a minority of patients with myocardial infarction are sent home on these agents.23 In a recent audit of post MRC Trial in the Elderly 62 48 infarct patients at the City Hospital, Birmingham, COOPE & WARRENDER – 30 EWPHE 65 – we found that only 41% of post infarct patients were SHEP 46 – sent home taking a beta-blocker despite the absence STOP-Hypertension – 33 of any contraindications. There was a slight trend for the cardiologists to use beta-blockers less fre- Beta-blockers for hypertension DG Beevers 809 Table 2 Results of the four clinical trials of the treatment of mild guideline committees of the British Hypertension hypertension which compared the effects of thiazide diuretics Society and the American Joint National Committee and beta-blockers on the development of fatal and non-fatal stroke, coronary heart disease (CHD) and all vascular deaths. For should seriously reconsider whether beta-blockers abbreviations see text. All significance values were in favour of have any role in the treatment of hypertension, the thiazide diuretics except possibly as a third line add-on drug in patients whose blood pressures remain uncontrolled Thiazide Beta-blockers Significance despite the combination of either an ACE inhibitor diuretics or an angiotensin receptor antagonist with either a calcium channel blocker or a thiazide diuretic. Their Number of trial participants MRC 4297 4403 – continued use in patients with coronary heart dis- MRC (Elderly) 1081 1102 – ease can only be justified in very high risk patients HAPPHY 3272 3297 – or those with persistent angina. IPPPSH 3172 3185 – TOTAL 11822 11987 – Postscript Stroke MRC 18 42 P Ͻ 0.01 It is of interest that since this review was written, MRC (Elderly) 16 21 NS and rejected by the British Medical Journal, an over- HAPPHY 41 32 NS view analysis by Professor Franz Messerli and col- IPPPSH 46 45 NS leagues has broadly come to the same conclusion.24 TOTAL 121 140 NS Neither of us were aware of the others work or opi- nions on the role of beta-blockers in hypertension. Coronary heart disease MRC 119 103 NS MRC (Elderly) 48 80 P Ͻ 0.01 Acknowledgements HAPPHY 116 132 NS IPPPSH 74 61 NS I am grateful for comments and advice from Drs J M TOTAL 354 376 NS Cruickshank and GYH Lip. The audit data from City Hospital were collected by Miss Fiona McHugh, 4th Vascular death year medical student. MRC 69 65 NS MRC (Elderly) 66 95 P Ͻ 0.05 HAPPHY 60 57 NS IPPPSH 56 45 NS References TOTAL 251 262 NS 1 Pritchard BNC, Gillam PMS. Use of propranolol (inderal) in the treatment of hypertension. Br Med J 1964; 2: 725–727. 2 Sever P et al. Management guidelines in essential quently than the general physicians. The reasons for hypertension: report of the second working party of this are uncertain but in a further 41% of cases the the British Hypertension Society. Br Med J 1993; 306: non-use of beta-blockade was due to contraindi- 983–987. cations including heart failure, heart block or 3 The sixth report of the Joint National committee on asthma. Prevention, Detection Evaluation and treatment of High Blood Pressure. Arch Iklern Med 1997; 157: Post infarct patient do have considerable mor- 2413–2446. bidity related to depression following their heart 4 Kendall MJ, Beeley L. Beta-adrenoceptor blocking attack and beta-blockers, by causing lethargy, tir- drugs: adverse reactions and drug interactions. Pharm- edness, sleep disturbance and impaired mentation, acol Ther 1982; 21: 351–369. must be regarded as an unattractive option. Their 5 Medical Research Council Working Party. Adverse use is however justified if the patient is at high risk reactions to bendrofluazide and propranolol for the or suffers from angina pectoris. treatment of mild hypertension. Lancet 1981; 2: 539– 543. 6 Fodor JG et al. A comparison of the side effects of aten- Conclusions olol and propranolol in the treatment of patients with hypertension. J Clin Pharmacol 1987; 27: 892–901. The beta-blocking drugs are less effective than other 7 MRC Working Party. Medical Research Council trial of agents in the majority of hypertensive patients and treatment of hypertension in older adults; principal they do not prevent myocardial infarction. They are results. Br Med J 1992; 304: 405–412. contraindicated in a large number of people and fre- 8 Coope J, Warrender TS. Randomised trial of treatment quently cause subtle but unpleasant side effects. In of hypertension in elderly patients in primary care. Br particular they can impair the quality of life of Med J 1986; 293: 1145–1151. hypertensive patients who are usually symptomless 9 Dahlof B et al. Morbidity and mortality in the Swedish until their doctor gives them treatment. It would Trial in old Patients with Hypertension (STOP appear that the Emperor has no clothes! Hypertension). Lancet 1992; 338: 1281–1285. With the advent of the ACE inhibitors, the newer 10 Amery A et al. Mortality and morbidity results from the European Working Party on High Blood Pressure calcium channel blockers and now the angiotensin in the Elderly trail. Lancet 1985; 2: 589–592. receptor antagonists, the clinician can use drugs 11 Avanzini F et al. Antihypertensive efficiency and tol- which do not cause these subtle and depressing side erability of different drug regimes in isolated systolic effects. These drugs must surely now be the pre- hypertension in the elderly. Eur Heart J 1994; 2: ferred option. In view of the above information, the 206–212. Beta-blockers for hypertension DG Beevers 810 12 SHEP Cooperative Research Group. Prevention of 19 Wilhelmsen L et al. Beta-blockers versus diuretics in Stroke by antihypertensive drug therapy in older per- hypertensive men. Main results from the HAPPHY sons with isolated systolic hypertension. JAMA 1991; trial. J Hypertens 1987; 5: 561–572. 265: 3255–3264. 20 MRS Working Party. MRC trial of treatment of mild 13 Fotherby MD, Harper GD, Potter JF. General prac- hypertension: Principal results. Br Med J 1985; 291: titioners’ management of hypertension in elderly 97–104. patients. Br Med J 1992; 305: 750–752. 21 Wikstrand J et al. Primary prevention with metoprolol 14 Hall WD. Pathophysiology of hypertension in blacks. in patients with hypertension. Mortality results from Am J Hypertens 1990; 3: 366S-371S. the MAPHY study. JAMA 1988; 259: 1976–1982. 15 Padfield PL et al. Is low-renin a stage in the develop- 22 Beta-Blocker Heart Attack Trial Research Group. A ment of essential hypertension or a diagnostic entity? randomised trial of propranolol in patients with acute Lancet 1975; 1: 548–561. myocardial infarction. II. Morbidity results. JAMA 16 Saunders E. Hypertension in backs. Med Clin North 1983; 250: 2814–2819. Am 1987; 71: 1013–1029. 23 ASPIRE Steering Group. A British Central Society sur- 17 Drayer JIM, Weber MA, Lipson JL, Megaffin BB. Differ- vey of the potential for the secondary prevention of ential effects of diuresis and beta-adrenoreceptor coronary disease: ASPIRE (Action on Secondary Pre- blockade during angiotensin-coverting enzyme inhibi- vention through Intervention to Reduce Events) princi- tors in patients with severe hypertension. J Clin Phar- pal results. Heart 1996; 75: 334–342. macol 1982; 22: 197–186. 24 Messerli FH, Grossman E, Goldbourt U. Are beta- 18 IPPPSH Collaborative Group. Cardiovascular risk and risk factors in a randomised trial of treatment based on blockers efficacious on first line therapy for hyperten- the beta-blocker oxprenolol: the International Prospec- sion in the elderly? JAMA 1998; 279: 1903–1907. tive Primary prevention Study in Hypertension (IPPPSH). J Hypertens 1985; 3: 379–392.